小土堆pytorch学习笔记（十五、优化器）

最新推荐文章于 2024-06-13 11:00:57 发布

Holy_cow

最新推荐文章于 2024-06-13 11:00:57 发布

阅读量80

点赞数

文章标签： pytorch 学习笔记深度学习 python

本文链接：https://blog.csdn.net/Holy_cow/article/details/132140331

版权

一、优化器

在这里插入图片描述
优化器是用来更新网络的参数的。
调试如下代码：

import torch
import torchvision.datasets
from torch import nn
from torch.nn import Sequential, Conv2d, MaxPool2d, Flatten, Linear
from torch.utils.data import DataLoader

dataset = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor(),
                                       download=True)
dataloader = DataLoader(dataset, batch_size=1)


class Cow(nn.Module):
    def __init__(self):
        super(Cow, self).__init__()
        self.model1 = Sequential(
            Conv2d(3, 32, 5, padding=2),
            MaxPool2d(2),
            Conv2d(32, 32, 5, padding=2),
            MaxPool2d(2),
            Conv2d(32, 64, 5, padding=2),
            MaxPool2d(2),
            Flatten(),
            Linear(1024, 64),
            Linear(64, 10)
        )

    def forward(self, x):
        x = self.model1(x)
        return x


loss = nn.CrossEntropyLoss()
cow = Cow()
optim = torch.optim.SGD(cow.parameters(), lr=0.01)
for data in dataloader:
    imgs, targets = data
    outputs = cow(imgs)
    result_loss = loss(outputs, targets)
    optim.zero_grad()
    result_loss.backward()
    optim.step()
    print(result_loss)

结果如下：
在这里插入图片描述

二、训练多个世代

代码：

for epoch in range(20):
    running_loss = 0.0
    for data in dataloader:
        imgs, targets = data
        outputs = cow(imgs)
        result_loss = loss(outputs, targets)
        optim.zero_grad()
        result_loss.backward()
        optim.step()
        running_loss = running_loss + result_loss
    print(running_loss)

定义了一个变量running_loss，对每个世代的损失求和，运行结果如下：

tensor(18701.4414, grad_fn=<AddBackward0>)
tensor(16160.4707, grad_fn=<AddBackward0>)
tensor(15474.5791, grad_fn=<AddBackward0>)
tensor(16005.3164, grad_fn=<AddBackward0>)
tensor(18091.4219, grad_fn=<AddBackward0>)
tensor(19951.2773, grad_fn=<AddBackward0>)
tensor(21966.7578, grad_fn=<AddBackward0>)
tensor(23601.1602, grad_fn=<AddBackward0>)
tensor(24501.1973, grad_fn=<AddBackward0>)
tensor(24851.4492, grad_fn=<AddBackward0>)
tensor(25541.3633, grad_fn=<AddBackward0>)
tensor(25367.8516, grad_fn=<AddBackward0>)
tensor(26323.0859, grad_fn=<AddBackward0>)
tensor(26919.5176, grad_fn=<AddBackward0>)
tensor(27632.3203, grad_fn=<AddBackward0>)
tensor(28924.0879, grad_fn=<AddBackward0>)
tensor(31648.1758, grad_fn=<AddBackward0>)
tensor(nan, grad_fn=<AddBackward0>)
tensor(nan, grad_fn=<AddBackward0>)
tensor(nan, grad_fn=<AddBackward0>)